Introduction to the Economics of Health Care

Introduction to the Economics of Health Care

 

 

The general objective of this course is to develop a set of analytical and conceptual tools that can be used to gain valuable insights into a host of health care issues and problems from an economic perspective. This topic takes the first step in accomplishing this important objective. In particular, this topic:

•                       introduces the discipline of health economics

•                       discusses resource constraints, trade-offs, efficiency, and equity

•                       highlights the state of the health economy in the United States and sets the stage for the material in the remaining topics

 

 

Like millions of Americans at some point in their lives, Joe awoke one night feeling a crushing weight on his chest. As the pain spread down his arm, he realized he was experiencing his worst dread: a heart attack. His wife, Angela, called the paramedics. While the ambulance rushed Joe to the hospital, she anguished over the kind of care he would receive. Angela’s anxiety starkly illustrates the basic questions any health care system faces:

 

1.                 Who should receive medical goods and services? Would a person like Joe receive care merely because he is a citizen, or would he receive care only if he worked for a large company that provides health insurance for its employees?

2.                 What types of medical goods and services should be produced? Should the most expensive tests (such as angiograms) be performed without regard to cost? What treatments (such as balloon angioplasties) should be provided?

3.                 What inputs should be used to produce medical goods and services? Should the hospital use high-tech medical equipment, a large nursing staff, or both?

 

All health care systems face questions such as these, but sometimes choose to answer them differently. When responding to health and health care questions, societies around the world take into account important moral, cultural, legal, economic, and other considerations. Addressing all of these concerns simultaneously and thoroughly is a daunting task, in part, because one concern often conflicts with another, but also because this task involves a substantial amount of time, effort, and knowledge. Indeed, the intellectual resource commitment would be so great that no one book could adequately cover all of the pertinent issues.

 

 

What Is Health Economics?

For many of you, this course provides your first exposure to the study of health economics. Perhaps the ongoing controversy regarding health care reform or the prospect of a career in the health care field motivated you to learn more about health economics. Or perhaps you need only three more credits to graduate. Whatever the reason, we are sure you will find health economics to be challenging, highly interesting, and personally rewarding.

The study of health economics involves the application of various microeconomics tools, such as demand or cost theory, to health issues and problems. The goal is to promote a better understanding of the economic aspects of health care problems so that corrective health policies can be designed and proposed. A thorough understanding of microeconomic analysis is essential for conducting sound health economics analyses. If you lack a background in microeconomics, don’t worry. This course is intended to help you learn and apply basic microeconomic theory to health economics issues. Before long, you will be thinking like a health economist!

The tools of health economics can be applied to a wide range of issues and problems pertaining to health and health care. For example, health economics analysis might be used to investigate why 25 of every 1,000 babies born in Turkey never reach their first birthday, whereas all but 3 of every 1,000 babies born in Japan live to enjoy their first birthday cake. The tools of health economics analysis might also be used to examine the economic desirability of a hotly contested merger between two large hospitals in a major metropolitan area. The burning question is: Will the merger of the two hospitals result in lower hospital prices due to overall cost savings or higher prices due to market power?

Health economics is difficult to define in a few words because it encompasses such a broad range of concepts, theories, and topics. The Mosby Medical Encyclopedia (1992, p. 361) defines health economics as follows:

 

Health economics… studies the supply and demand of health care resources and the impact of health care resources on a population.

 

Notice that health economics is defined in terms of the determination and allocation of health care resources. This is logical, because medical goods and services cannot exist without them. Even health care services produced in the home, such as first aid (therapeutic services) or home pregnancy tests (diagnostic services), require resources. Health care resources consist of medical supplies, such as pharmaceutical goods, latex rubber gloves, and bed linens; personnel, such as physicians and lab assistants; and capital inputs, including nursing home and hospital facilities, diagnostic and therapeutic equipment, and other items that provide medical care services. Unfortunately, health care resources, like resources in general, are limited or scarce at a given point in time, and wants are limitless. Thus, trade-offs are inevitable and a society, whether it possesses a market-driven or a government-run health care system, must make a number of fundamental but crucial choices. These choices are normally couched in terms of four basic questions, discussed next.

 

The Four Basic Questions

As just noted, resources are scarce. Scarcity means that each society must make important decisions regarding the consumption, production, and distribution of goods and services as a way of providing answers to the four basic questions:

1.                   What mix of nonmedical and medical goods and services should be produced in the macroeconomy?

2.                   What mix of medical goods and services should be produced in the health economy?

3.                   What specific health care resources should be used to produce the chosen medical goods and services?

4.                   Who should receive the medical goods and services that are produced?

 

How a particular society chooses to answer these four questions has a profound impact on the operation and performance of its health economy.

The first two questions deal with allocative efficiency: What is the best way to allocate resources to different consumption uses? The first decision concerns what combination of goods and services to produce in the overall economy. Individuals in a society have unlimited wants regarding nonmedical and medical goods and services, yet resources are scarce. As a result, decisions must be made concerning the best mix of medical and nonmedical goods and services to provide, and this decision-making process involves making trade-offs. If more people are trained as doctors or nurses, fewer people are available to produce nonmedical goods such as food, clothing, and shelter. Thus, more medical goods and services imply fewer nonmedical goods and services, and vice versa, given a fixed amount of resources.

The second consumption decision involves the proper mix of medical goods and services to produce in the health economy. This decision also involves trade-offs. For example, if more health care resources, such as nurses and medical equipment, are allocated to the production of maternity care services, fewer resources are available for the production of nursing home care for elderly people. Allocative efficiency in the overall economy and the health economy is achieved when the best mix of goods is chosen given society’s underlying preferences.

The third question – what specific health care resources should be used? – deals with production efficiency. Usually resources or inputs can be combined to produce a particular good or service in many different ways. For example, hospital services can be produced in a capital-or labor-intensive manner. A large amount of sophisticated medical equipment relative to the number of patients served reflects a capital-intensive way of producing hospital services, whereas a high nurse-to-patient ratio indicates a labor-intensive process. Production efficiency implies that society is getting the maximum output from its limited resources because the best mix of inputs has been chosen to produce each good.

 

 

Production and Allocative Efficiency and the Production Possibilities Curve

The most straightforward way to illustrate production and allocative efficiency is to use the production possibilities curve (PPC). A PPC is an economic model that depicts the various combinations of any two goods or services that can be produced efficiently given the stock of resources, technology, and various institutional arrangements. Figure 1-1 displays a PPC. The quantities of maternity services, M, and nursing home services, N, are shown on the vertical and horizontal axes, respectively. We assume society has already made its choice between medical and nonmedical goods. Points on the bowed-out PPC depict the various combinations of maternity and nursing home care services that can be efficiently produced within a health economy assuming the amounts of health care resources and technology are fixed at a given point in time.

 

 

The PPC shows the trade-off between any two goods given a fixed stock of resources and technology. Any point on the PPC, such as points A through E, reflects efficiency because units of one good must be given up to receive more of the other. A point in the interior, such as F, reflects inefficiency because more of one good can be attained without necessarily reducing the other. A point outside the PPC, such as G, is not yet attainable but can be reached with an increase in resources or through institutional or technological changes that improve productivity.

 

Every point on the PPC implies production efficiency, since all health care resources are being fully utilized. For example, notice points A, B, C, D, and E on the PPC. At each of these points, medical inputs are neither unemployed nor underemployed (for example, a nurse involuntarily working part time rather than full time) and are being used in the most productive manner so that society is getting their maximum use. If a movement along the curve from one point to another occurs, units of one medical service must be forgone to receive more units of the other medical service.

Specifically, assume the health economy is initially operating at point C with M_C units of maternity care services and N_C units of nursing home services. Now suppose health care decision makers decide that society is better off at point D with one more unit of nursing home services, N_D – N_C. The movement from point C to point D implies that M_C – M_D units of maternity care services are given up to receive the additional unit of nursing home services. Because medical resources are fully utilized at point C, a movement to point D means that medical inputs must be drawn or reallocated from the maternity care services market to the nursing home services market. As a result, the quantity of maternity care services must decline if an additional unit of nursing home services is produced. The forgone units of maternity care services, M_C – M_D, represent the opportunity cost of producing an additional unit of nursing home services. As economists are fond of reminding noneconomists, ‘There is no such thing as a free lunch!” Generally, opportunity cost is the value of the next best alternative that is given up.

The bowed-out shape of the PPC implies that opportunity cost is not constant but increases with a movement along the curve. Imperfect substitutability of resources is one reason for this so-called law of increasing opportunity cost. For example, suppose the nursing home services market expands downward along the PPC. To produce more nursing home services, employers must bid resources away from the maternity care services market. Initially, the least productive inputs in the maternity care services market are likely to be bid away, because they are available at a lower cost to nursing home employers. Consequently, very few maternity care services are given up at first. As the nursing home services market continues to expand, however, increasingly productive inputs in the maternity care services market must be drawn away. The implication is that society gives up ever-increasing units of maternity care services. Thus, the law of increasing opportunity cost suggests that ever-increasing amounts of one good must be given up to receive successively more equal increments of another good.

If medical inputs are not fully utilized because some inputs are idle or used unproductively, more units of one medical service can be produced without decreasing the amount of the other medical service. An example of an underutilization of resources is indicated by point F in the interior of the PPC. At point F, the health care system is producing only M_F units of maternity services and N_F units of nursing home services. Notice that by moving to point B on the PPC, both maternity care services and nursing home services can be increased without decreasing the other. The quantities of both goods increase only because some resources are initially idle or underutilized at point F. Health care resources are inefficiently employed at point F.

A point outside the current PPC, such as G, is attainable in the future if the stock of health care resources increases; a new, productivity-enhancing technology is discovered; or various economic, political, or legal arrangements change and improve productive relationships in the health economy. If so, the PPC shifts out and passes through a point like G. For example, technological change may enable an increased production of both maternity and nursing home services from the same original stock of health care resources. Alternatively, a greater quantity of maternity and nursing home services can be produced and the PPC shifts outward if more people enter medical professions (possibly at the expense of all other goods and services).

Production efficiency is attained when the health economy operates at any point on the PPC, since medical inputs are producing the maximum amount of medical services and no unproductive behavior or involuntary unemployment exists. Allocative efficiency is attained when society chooses the best or most preferred point on the PPC. All points on the PPC are possible candidates for allocative efficiency. The ideal, or optimal, point for allocative efficiency depends on society’s underlying preferences for the two medical services.

Of course, the real world is much more complex than the example depicted by the PPC. Rather than only two goods, an unimaginable number of goods and services are produced in a society. The PPC is a model because it offers a simplification of reality. As pointed out in more depth in appendix A1, models are useful in the field of economics because they serve as conceptual devices or tools for organizing our thoughts about a topic. The PPC provides a good example of a simple but powerful model because it sheds light on a number of important lessons including: (1) the all-important economic role of scarcity; (2) the significance of economic choices; (3) the costs of inefficiency; and (4) how growth takes place in an economy.

 

 

The Distribution Question

The answer to the fourth question – who should receive the medical goods and services? – deals with distributive justice or equity. It asks whether the distribution of services is equitable, or fair, to everyone involved. In practice, countries around the world have chosen to address this medical care distribution question in many different ways.

When thinking about the distribution question, it is sometimes useful to consider two theoretically opposite ways of distributing output: the pure market system and a perfect egalitarian system. Goods and services are distributed in a pure market system based solely on each person’s willingness and ability to pay because decisions concerning the four basic questions are answered on a decentralized basis within a system of markets. That is, goods and services are distributed, or rationed, to only those people who are both willing and able to purchase them in the marketplace. Because people face an incentive to earn income to better afford goods and services in a pure market system, they tend to work hard and save appropriately for present and future consumption. Consequently, productive resources tend to be allocated efficiently in a pure market system. In other words, the incentives associated with a pure market system typically mean that the economy operates on the PPC.

In many cases, differences in ability to pay among individuals reflect that some have consciously chosen to work harder and save more than others. Unfortunately, differences in ability to pay may also indicate that some people have less income because of unfortunate life circumstances such as a mental, physical, or social limitation. Regardless of the specific reason, it follows that people without sufficient incomes face a financial barrier to obtaining goods and services in a pure market system in which price serves as a rationing mechanism. Given income disparities, some people may be denied access to needed goods and services. Consequently, the pure market system is typically viewed as inherently unfair by many when it comes to the distribution of important goods and services such as health care.

In direct contrast, a central committee, such as a federal or subnational unit of government, may answer the distribution question by ensuring everyone receives an equal share of goods and services. In an egalitarian system of this kind, everyone has access to the same goods and services without regard to income status or willingness to pay. Therefore, no one is denied access to needed goods and services. But an incentive may exist for people to choose to work and save less because the consumption decision is divorced from the distribution of earned income. Because of this inefficient allocation of resources, fewer goods and services may be available for distribution in an egalitarian system. In this case, the economy may operate inside the PPC.

In practice, most countries have adopted a mixed distribution system, with the reliance on central versus market distribution varying by degree across countries. For example, in the United States, many goods and services are distributed by both the market and the government. The food stamp, temporary assistance for needy families, and Medicaid programs represent some of the many policies adopted by the U.S. government to redistribute goods and services. Some people applaud these programs, whereas others argue that they worsen both efficiency and equity. They argue that efficiency and equity are compromised when those who choose to commit fewer resources to production are rewarded through redistributive programs and productive individuals are penalized via taxation. The efficiency and equity implications of various redistributive policies are constantly debated in the United States and elsewhere. In the context of health care, the consequence of this debate regarding distribution might determine who lives and who dies. For this reason, among others, more discussion on the redistributive function of government is taken up in Topics 9 and 10.

 

 

Implications of the Four Basic Questions

Given a scarcity of economic resources, a society generally wishes to produce the best combination of goods and services by employing least-cost methods of production. Trade-offs are inevitable. As the PPC illustrates, some amount of one good or service must be given up if the production and consumption of another good or service increases. As a result, each society must make hard choices concerning consumption and production activities because scarcity exists. Choices may involve sensitive trade-offs, for example, between the young and the old, between prevention and treatment, or between men (prostate cancer) and women (breast cancer).

In addition, some individuals lack financial access to necessary goods and services such as food, housing, and medical care. Because achieving equity is a desirable goal, a society usually seeks some redistribution of income. Normally, the redistribution involves taxation. However, a tax on labor or capital income tends to create a disincentive for employing resources in their most efficient manner. This point is discussed in more detail in Topic 9. Inefficient production suggests that fewer goods and services are available in the society (production inside the PPC). Thus, a trade-off often exists between equity and efficiency goals, and, consequently, hard choices must be made between the two objectives. The design of a nation’s health care system normally reflects the way the society has chosen to balance efficiency and equity concerns.

 

 

Taking the Pulse of the Health Economy

A health economy, like a macroeconomy, involves the production and consumption of goods and services and the distribution of those goods to consumers. A health economy differs from a macroeconomy because it distinctly considers production, consumption, and distribution activities that directly relate to population health. More will be said about that difference in topics 2 and 4. Another difference concerns the way in which economists take the pulse of the macroeconomy and health economy. While economists are really concerned with efficiency and equity, the unemployment, inflation, and gross domestic product growth rates are also considered when gauging the performance of a macroeconomy. If you recall from ECON 100, gross domestic product (GDP) captures the total market value of all goods and services produced in an economy during a particular period.

For a health economy, the analogous performance indicators are the components that make up the so-called three-legged stool of medical care: costs, access, and quality. Again, although health economists are more concerned about efficiency and equity, many often use some variation of the three-legged medical stool to gauge the performance of a health economy. We discuss and provide some historic and contemporary data for each of these components in the following sections. The discussioot only introduces the various legs of the medical stool, but also motivates and acts as a roadmap for the remaining material in this course.

 

 

Medical Care Costs

Although the topic of medical care costs is taken up more formally in Topic 7, recall from our earlier discussion that medical care resources, like resources in general, are scarce at a given point in time. It follows that an opportunity cost, or a price, is associated with each and every medical care resource because of scarcity. Thus, we can think of medical care costs as representing the total opportunity costs when using various societal resources such as labor and capital to produce medical care rather than other goods and services.

Each year since 1960, actuaries at the Centers for Medicare and Medicaid Services (CMS) have collected and reported data on the uses, sources, and costs of medical care in the United States. The data can be compared across various industries in the health care sector, like hospital, physician, and nursing home services, examined in a particular year, or tracked over time. Funding sources including consumers, insurers, or government can also be examined for various types of medical care, and over time. Hence, the CMS data yield important insights with respect to how health care funds are used, where the funds come from, and how much money in total is spent on medical care in the United States.

 

 

Uses of Medical Funds

Figure 1-2 provides a percentage breakdown of the uses of health care funds in 2006. These statistics offer insight into the mix of medical goods and services actually produced and consumed in the U.S. health economy. Recall that the second basic question is “what mix of medical care ‘should be’ produced.” Also recall that more of one type of medical care means less of the others for a given size of the medical care pie.

 

 

According to the figure, 31 percent of medical care funds is spent on hospital services. The “big ticket” nature of hospital services should not be too surprising. Acutely ill individuals typically stay for a fairly long time in a hospital at some point in time. Physician services make up the next largest use of funds with 21 percent of the total. The dominant role of physicians makes sense because they are primary care gatekeepers and patients must often first pass through them before accessing other types of medical care, including hospitals and prescription drugs. In addition, specialty physicians, such as heart surgeons, provide important services that maintain, improve, and extend human lives. Their reimbursement reflects, in part, the value placed on remaining healthy, which is discussed in Topic 3.

Collectively, hospital and physician services account for more than half of all health care spending, not only in 2006 but over time as well. We will learn more about the structure, conduct, and performance of the physician and hospital services markets in Topics 12 and 13, respectively. Finally, prescription drugs (10 percent), nursing home care (6 percent), dental care (4 percent), and home health care (3 percent) represent four other major areas where medical care funds are directly spent on patient care.

 

 

Sources of Medical Funds

 

The percentage of medical funds coming directly from consumers, private insurers, and government are shown in Figure 1-3. We emphasize the word directly because all funds ultimately come from the consumer in the form of out-of-pocket payments, premiums, and/or taxes. In 2006, 54 percent of all funds spent oational health care came from the private sector, down from approximately 76 percent in 1960. The bulk of this decrease took place in the mid-1960s when two public health insurance programs – the Medicare and Medicaid programs – were first introduced. Since 1990, the share of national health expenditure emanating directly from the private sector has dropped slightly from 59 percent.

 

 

The mix between private insurance and out-of-pocket payments has also changed in recent years. In particular, private insurance has expanded its role as a source of funds and substituted greatly for out-of-pocket payments. In 1980, for example, private health insurance provided funds for 29 percent of all health care costs in the nation and out-of-pocket payments provided another 17 percent. By 2006, slightly over one-third of national health care expenditures came from private insurers while consumers out-of-pocket payments fell to 12 percent. The greater reliance on private insurance funding reflects both a greater number of individuals and more types of medical care (e.g., pharmaceuticals and dental) covered by medical insurance. Business payments to provide health care services directly to employees, philanthropic sources, private construction, and nonpatient revenue sources (such as revenues from hospital gift shops), help to account for the remaining 8 percent of all private funds in 2006.

Figure 1-3 also shows that 46 percent of all national health spending in 2006 came from the government. Most of the government funds were spent by the Medicare and Medicaid public health insurance programs. Given that the government funds less than half of all health care spending in the nation, the United States is often looked upon as possessing a privately financed health care system. However, Woolhandler and Himmelstein (2002) offer an alternative view of the relative share of health care spending financed through private and public sources. In particular, they scrutinize the method used by CMS to measure government spending in the national health accounts and show that the government has much more responsibility than the private sector with respect to financing the U.S. health care system.

Woolhandler and Himmelstein explain that CMS includes only direct purchasing of medical care for programs such as Medicare, Medicaid, and government-owned hospitals in its measure of government spending. Consequently, public employee benefits, such as those through the Federal Employees Health Benefits Program and various state employee health insurance programs, are missing from CMS’s reported figures. Although the government supports these public health insurance programs with tax financing, private insurers administer the program on behalf of the government and are responsible for writing the actual checks. In addition, the authors point out that employer-sponsored health insurance premiums are exempted from various federal, state, and city taxes. (We take this up later in Topic 6.) Thus, the government also implicitly helps to finance employer-sponsored health insurance through these tax preferences.

To get a better idea about the extent to which health insurance is tax financed, the authors add together the direct purchasing of medical care by government with expenditures on public employee health benefits that are tax-financed but administered by the private sector plus the value of the health insurance premium tax preference. Woolhandler and Himmelstein report that the direct spending of government equaled 45 percent of all health care costs, while public employee benefits accounted for another 5.4 percent, and the tax subsidy for health insurance premiums amounted to an additional 9.1 percent. Thus, government, at all levels, was responsible for financing nearly 60 percent of all health care costs in the United States. Thus, one might rightfully argue that, similar to other countries around the world, the government largely finances the health care system in the United States.

These estimates of Woolhandler and Himmelstein are certainly provocative. They show that tax financing represents the major source of health care funds in the United States. Indeed, tax financing accounts for an even greater share of health care costs, considering that not-for-profit health care organizations such as hospitals, behavioral health care organizations, and nursing homes are also granted preferences on income, property, and sales taxes. (We also take this up in later topics.) How these highly credible estimates are interpreted and used in future policy discussions concerning health care reform will be interesting to see.

 

 

Amount of Medical Care Spending

Only someone living in entire seclusion, perhaps a World War II Japanese soldier hiding somewhere on a Pacific island or someone raised in a nuclear fallout shelter of the 1950s, would be unaware of the situation involving medical care costs in the United States. One of the authors was stationed in Guam during the Vietnam conflict. A World War II Japanese soldier was rumored to be hiding on the island. View the movie Blast from the Past starring Brendan Fraser to learn how growing up in a fallout shelter can affect one’s knowledge of current events. Indeed, it seems that not a day goes by without a radio, television, or popular press commentator pointing, with much alarm, to the high and continually rising costs of health care. There is certainly no need to dispute those facts. According to CMS figures, the United States spent $2.1 trillion on health care or slightly over $7,000 per person in 2006. Compare that to the similar figures of $26.9 billion and $141 dollars in 1960.

These figures are potentially alarming because trade-offs may be involved. That is, the PPC tells us that high health care costs translate into lower amounts of other goods produced and consumed. Certainly, high health care costs could reflect more and better medical care, but high spending may also involve the sacrifice of other equally important goods and services like food, clothing, and shelter. However, the productive capacity of the U.S. health economy has changed over time – the situation may not be as bleak as the statistics show. For example, the economy may now possess more labor and capital resources and productivity-improving technologies. Thus, the PPC has likely shifted out and therefore more of one good or service can be produced without sacrificing the others.

One way of controlling for differences in the underlying productive capacity of an economy or economies is by dividing, in this case, the amount of health care spending by GDP. Greater productive capacity, resulting from higher amounts of resources and better technology, generally means a larger level of GDP and therefore more goods and services in general. With that notion in mind, Figure 1-4 shows health care spending as a percentage of GDP from 1960 to 2006.

Figure 1-4 shows that health care spending as a percentage of GDP has grown tremendously over time in the United States. Standing at 5.2 percent in 1960, that same ratio of health care spending to GDP is now about 16 percent, which means instead of spending $1 out of every $20, we now spend $1 out of every $6 on health care. However, even the rising percentage of GDP devoted to health care does not necessarily indicate other goods and services have been sacrificed. The GDP of $13 trillion in 2006 is much greater than the GDP of $526 billion in 1960. Given the health care spending to GDP ratios in the two years, spending on all other goods amounted to nearly $11 trillion in 2006 compared to $495 billion in 1960. Simply put, the greater productive capacity of the U.S. economy allowed for greater amounts of both health care and all other goods to be produced.

 

 

In fact, productivity-enhancing technologies in the rest of the economy may have freed up resources for use in the health economy where the labor intensity of medical services doesn’t allow us much productivity improvement. Of course, the relative mix of goods has certainly favored the health care sector since 1960.

Figure 1-4 also shows that health care spending has not increased at the same continuous rate throughout the years. For example, health care spending grew more quickly relative to GDP prior to the 1990s. In contrast, notice that after the 1990s, the ratio of health care spending to GDP remained relatively stable during the 1993 to 1999 period. The ratio of health care costs to GDP has also remained fairly constant since 2003. To examine the efficiency consequences of medical spending we must consider the benefits of medical goods and services in addition to their costs. That topic is taken up in Topic 3.

Policy makers continue to debate the cause and desirability of rising health care costs in the United States and in other countries. Some argue that the U.S. health care system contains a lot of production inefficiency that can and should be squeezed out. Others point out that the benefits from health care more than compensate for the costs. Much of this debate is covered in various topics of this book. It shouldn’t be too surprising that health economists are heavily involved in this debate. In fact, they often draw upon the tools that can be learned in this book when trying to make some sense of health care spending and the health care economy. The structure of a health care system certainly plays a role so that topic is taken up in Topic 4. The material in Topics 5, 6, 7, and 8 also add to our understanding of health care costs and how consumers, providers, insurers, markets, government, and economic incentives help to shape health care spending.

 

 

Medical Care Access

Medical care access, another leg of the medical stool, relates to the distribution question. That is: Does everyone have reasonable access to medical care on a timely basis? Timely access is often measured by the percentage of individuals with health insurance. For most people the cost of catastrophic care, such as organ transplants and cariovascular surgery, lies beyond their financial means. But, as explained fully in Topic 6, for a relatively small payment or premium, insurance provides access to high-cost, life-saving interventions if and when people experience severe illnesses. Thus, health insurance may be an important factor in terms of ensuring timely access to medical care. Figure 1-5 offers some information on the percentage of people without health insurance in the United States since 1940.

Before discussing the data in Figure 1-5, it should be noted that the health insurance product has changed considerably over time. Prior to the 1970s most people purchased only hospital insurance. Today people purchase health insurance for other types of medical care, as mentioned previously. Also, the amount of medical care expenditures covered by insurance has increased over the years. Thus, for the sake of consistency, it may be best to think of Figure 1-5 as showing the percentage of the U.S. population without hospital insurance.

In any case, the data in the figure show that great strides have been taken in terms of more people insured in the United States. In 1940, only 10 percent of the U.S. population possessed health insurance purchased almost entirely in the private marketplace. Even before public health insurance programs, beginning with the Medicare and Medicaid Acts of the mid-1960s, many people began purchasing private health insurance in the United States following the 1940s. By 1975 the uninsured rate in the United States dipped to about 13 percent because of both private purchases and public expansions.

 

 

 

 

However, beginning around 1980, further persistent declines in the uninsured rate have not materialized. In 2006 the uninsurance rate in the United States stood at nearly 16 percent. While we take up the causes, types, and social costs of uninsurance in Topics 6 and 11, it suffices to note that a sizeable percentage of the U.S. population lacks timely access to medical care because of their uninsured status. In addition, severe racial disparities exist with respect to uninsured status, as noted in Topic 11. Clearly, these are two additional areas where the tools of health economies are needed to shed better light and bring about improvements in the health economy and society.

 

 

Medical Care Quality

The final leg of the medical stool we consider is medical care quality. As discussed more fully in Topic 2, quality represents a complex and multidimensional concept. In keeping with the other two legs of the medical stool we confine our discussion to a single measure of quality that is easily understandable and important from a societal point of view, and for which data can be obtained over time for comparative purposes. The chosen measure is the infant mortality rate that tells us the number of children below one year of age that died as a percentage of all live births in that same year. The infant mortality rate for the United States from 1960 to 2006 is reported in Figure 1-6.

 

 

Like the uninsured rate, the infant mortality rate has improved significantly over time in the United States falling from a height of over 25 infant deaths per 1,000 live births in 1960. Although it stands to reason that rising health care spending and increased insurance coverage contributed to the decline, Topic 2 discusses the theoretical framework and empirical findings regarding the many factors influencing health status outcomes such as infant mortality. Despite the vast improvements that have taken place over time, Figure 1-6 suggests that nearly 7 out of every 1,000 live babies in the United States do not live beyond 1 year of age. Also, the United States lags far behind when compared to other industrialized countries like Belgium, France, Italy, Japan, and the United Kingdom, which have infant mortality rates below 5 deaths per 1,000 live births. Finally, the figure does not capture the vast variations in health outcome measures, such as infant mortality, among different income, racial, and ethnic groups. Once again, the tools of health economics can prove useful for analyzing health outcomes and proposing ways of improving societal health.

 

 

A Note on the Relation between System Structure and Performance

Many theories and empirical findings pertaining to health economics are introduced and developed in this text. Sometimes theories and empirical findings are of interest for their own sake, particularly for academicians such as the authors. But the main reason for their introduction and development is that we wish to obtain a better grasp of the operation and performance of the real-world health economy around us. If the health economy does not perform in a socially efficient and equitable manner, then we would hope that solutions could be proposed and policies could be changed to alter that undesirable performance.

 

 

An understanding of the link between structure and performance is essential when crafting new policies. Structure plays a role in determining how people behave or conduct themselves in the health economy. Figure 1-7 shows the complex interaction between structure and performance. A health economy is structured in a particular way, and this health economy structure is discussed in great detail in Topic 4. Structure shows up in the ways various organizations are designed in terms of their size and scope, the mix of market activities and government involvement in the health economy, and financing and reimbursement mechanisms, among others.

This underlying structure helps to establish the prevailing incentives in a health economy and thereby influences how people, organizations, and government itself, behave. If incentives are distorted because of structural defects, then suboptimal performance likely results in terms of inefficient and inequitable outcomes. Given the suboptimal performance, solutions can be proposed and public policies can be designed to remedy the situation. In particular, policies can be changed to either indirectly affect behavior through a restructuring of the system or directly by introducing conduct remedies.

Just about every topic in the book addresses an issue where incentives are discussed or public policy plays a role. As mentioned previously, health economists are most interested in the efficiency of outcomes because resources are scarce. Unfortunately, efficiency is often difficult to gauge or measure in practice. An alternative is to design a theoretical benchmark where efficiency can be attained; then compare the real world, in terms of the existing incentives because of its structure, to that theoretical benchmark. Our benchmark for allocative efficiency (the point at which marginal social benefit equals marginal social cost) is developed in Topic 3. This benchmark is expanded upon in Topic 8. The most discussion concerning public policy shows up in Topics 9, 10, and 16.

 

 

Summary

Health economics is concerned with the determination and allocation of health resources and distribution of medical services in a society. Because resources are scarce, society must determine what amounts of medical services to produce, what kinds of medical services to produce, what mix of health care resources should be used, and who should receive the output of health care services. Answering these four basic questions involves tough trade-offs.

A health economy, like a macroeconomy in general, can be analyzed with respect to its performance. We discussed how the health economy can be assessed with regard to medical care cost, access, and quality and learned that the tools of health economics can and will be used to explore more thoroughly these components of the three-legged medical stool in subsequent topics of this book. Controlling medical costs, access, and quality also involves trade-offs.

Finally, economic analysis can help us better understand the causes of problems relating to health and health care. The tools and concepts of health economics can also be used to find solutions and offer public policy prescriptions. The public policy prescriptions may involve structural and/or conduct remedies.

 

 

Review Questions and Problems

1.                Draw a bowed-out PPC with an aggregate measure of medical services, Q, on the horizontal axis and an aggregate measure of all other goods (and services), Z, on the vertical axis. Discuss the implications of the following changes on the quantities of medical services and all other goods.

A.     A movement down along the curve.

B.      A movement from the interior of the curve to a northeasterly point on the curve.

C.     An increase in the quantity of labor in the economy.

D.     A technological discovery that increases the production of Z.

         If it were your choice, where would you choose to produce on the PPC? Why?

2.                Congratulations! Upon graduating you accept a well-deserved job with XER Consulting. Your first job involves a consulting gig with a state subcommittee on health care issues. The senate health care subcommittee is considering the expansion of two existing public health programs. One program concerns additional funding for nursing homes around the state. The other program involves additional funding for community health centers around the state. In both cases the funding is supposed to be used to attract more nurses for expansion purposes. Your job involves the following four tasks:

A.     Draw and use a production possibilities curve to graphically show and verbally explain to the subcommittee members the opportunity cost at a point in time of expanding any one of the programs, assuming that both of them are initially operating efficiently. Be sure to correctly label the axes and all points. Refer to the points on the graph in your explanation.

B.      Use the production possibilities curve to graphically show and verbally explain how one or both programs could be expanded at a lower opportunity cost if some inefficiency or slack initially exists in the overall public health system. Refer to various points on the graph in your explanation.

C.     Use the production possibilities curve to graphically show and verbally explain how both programs could be expanded at a lower opportunity cost if growth is expected for the public health care system. Refer to points on the graph in your explanation.

D.     Verbally explain to the subcommittee members what factors might cause the public health care system to grow.

3.                Identify the so-called three legs of the medical stool. Explain how trade-offs might take place among the three legs. If you had to choose one of the three to improve upon at the neglect of the others, which would you choose? Why?

4.                Does the U.S. health care system possess a privately or publicly financed health care system? Explain.

5.                What are two major uses of medical funds? How do the two major uses relate to the four basic questions?

6.                At this point in the book, do you think the United States spends too much on medical care? Explain your reasoning using the PPC.

7.                Explain the change in the percentage of the U.S. population with health insurance from 1940 to 1980. Can you think of any economic factors that may have caused that change? Explain the change in the percentage insured since 1980.

8.                Explain the change in the infant mortality rate (IMR) in the United States since 1960. Do you think the IMR is too high in the United States? Why? What is the implication of a reduction in the IMR if we treat infant mortality rate reductions as one good on one axis of the PPC and all other goods on the other axis? What is the implication of an IMR reduction if we assume some production inefficiency initially exists in the U.S. health care system? Why?

9.                In your own words, explain the general link between system structure, performance, and policy.

 

 

References

 

The Mosby Medical Encyclopedia. New York: C. V. Mosby, 1992.

Organization for Economic and Cooperative Development. OECD Health Data 2006. October 2006.

Santerre, Rexford E. “Tracking Uninsurance and Inflation in the U.S. Health Economy.” University of Connecticut, Center for Healthcare & Insurance Working Paper 2007-05 (September 2007).

Woolhandler, Steffie, and David U. Himmelstein. “Paying for National Health Insurance – and Not Getting It.” Health Affairs 21 (July/August 2002), pp. 88-98.

 

 

Appendix 1: Economic Models and Empirical Testing

 

Health economics can be considered as both a social science and a science. In fact, economics, of which health economics is a subdiscipline, touches upon history, psychology, sociology, philosophy, mathematics, and statistics. As a social science, the field of health economics studies people in their everyday lives and addresses issues such as obesity, alcohol abuse, and abortion. As a science, health economics offers testable hypotheses. For example, a health economist might explore empirically if people purchase more whiskey or fast food when their prices decline – the so-called law of demand. In either case, models and empirical methods are used in health economics. This appendix offers an introduction to both of these tools of health economic analysis.

 

Economic Models

As mentioned earlier, the PPC is an example of an economic model. Models are abstractions of reality and are used in economics to simplify a very complex world. Economic models can be stated in descriptive (verbal), graphical, or mathematical form. Usually an economic model like the PPC describes a hypothesized relation between two or more variables. For example, suppose the hypothesis is that health care expenditures, E, are directly (as opposed to inversely) related to consumer income, Y That hypothesis simply means that expenditures on health care services tend to rise when consumer income increases. Mathematically, a health care expenditure function can be stated in general form as

 

(A1-1)                                              E = f (Y).

 

Equation A1-1 implies that health care spending is a function of consumer income. In particular, health care expenditures are expected to rise with income.

An assumption underlying economic models is that all factors, other than the variables of interest, remain unchanged. For example, our hypothesis that health care expenditures are directly related to income assumes that all other likely determinants of health care spending, such as prices, tastes, and preferences, stay constant. As another example, notice in the previous analysis that the stocks of resources and technology are held constant when constructing the PPC. Indeed, economists normally qualify their hypotheses with the Latin phrase ceteris paribus, meaning “all other things held constant.” By holding other things constant, we can isolate and describe the pure relation between any two variables.

The expenditure function in Equation A1-1 is expressed in general mathematical form, but a hypothesis or model is often stated in a specific form. For example, the following equation represents a linear expenditure function for health care services:

 

(A1-2)                                     E = a + bY,

 

where a and b are the fixed parameters of the model. This equation simply states that health care expenditures are directly related to consumer income in a linear (rather than nonlinear) fashion. Mathematically, the parameter a reflects the amount of health care expenditures when income is zero, whereas b is the slope of the expenditure function. The slope measures the change in health care expenditures that results from a one-unit change in income, or AE/AY.

For example, let us assume the parameter a equals $1,000 per year and b equals one-tenth, or 0.1. The resulting health care expenditure function is thus

 

(A1-3)                                    E = 1,000 + 0.1 Y.

 

Equation A1-3 implies that health care expenditures rise with income. In fact, the slope parameter of 0.1 suggests that each $1,000 increase in consumer income raises health care spending by $100.

The health care expenditure function in Equation A1-3 is represented graphically in Figure A1-1. Yearly consumer income per household is shown on the horizontal axis, and annual health care spending per household is shown on the vertical axis. According to the function, health care spending equals $3,000 when household income is $20,000 per year. Consumers earning $50,000 per year spend $6,000 per year on health care services. Note that the expenditure function clearly represents our hypothesis concerning the direct relation between income and health care spending.

Now suppose some other determinants of health care expenditures change. Although this assumption violates our implicit ceteris paribus condition, we can incorporate changes in other factors into the health care expenditure model fairly simply. For example, suppose people generally become sicker than before, perhaps because households have become older on average. Obviously, this change tends to increase health care spending. Assuming that the “aging” effect influences only the intercept term and not the value of the slope parameter, the expenditure function shifts upward by the yearly increase in health care spending due to the aging population. Figure A1-2 shows an example of this effect.

Yearly medical costs are assumed to increase by $500 for the typical household. Thus, the health care expenditure function shifts upward at each level of income by $500 to E₁. If the aging effect also influences the percentage of additional income that people spend on health care services, the slope of the function changes as well. An increase (decrease) in the marginal propensity to spend out of income raises (lowers) the slope and rotates the expenditure function to the left (right). Problem 2 at the end of the topic asks you to complete an exercise of this type.

As you can see, a model, such as this expenditure function or the PPC, is useful because it helps simplify an otherwise complex world. We can better and more easily understand the relation among key variables. Models are also useful because they often offer valuable insights into the necessity or relative effectiveness of various public policies. For example, we saw from the PPC that policy changes typically involve trade-offs that public policy makers should heed.

 

 

 

According to the expenditure function, health care spending increases with income. For example, health care spending equals $3,000 when household income equals $20,000 per year and $6,000 when household income equals $50,000 per year.

 

In the case of our health care expenditure function, suppose that some government agency, such as the U.S. Government Accountability Office or Congressional Budget Office, determines that $4,000 of annual household spending on health care is necessary to maintain the health of family members in the typical household. Further suppose that a study by this same government agency finds that our health care expenditure model, as reflected in Equation A1-3, represents the true relation between household income and health care spending. If so, our model suggests that households with incomes less than $30,000 tend to spend less than the necessary amount on health care. The government might use this information to determine the subsidy needed at each level of family income to reach the targeted amount of $4,000. For example, a household with $10,000 of income would require a $2,000 subsidy to reach the targeted amount of health care spending whereas a household with $28,000 would need only $200.

 

 

Yearly health care spending is assumed to increase by $500 for a reason other than a change in income. Thus, the expenditure function shifts upward at each level of income by $500 to E₁.

Consequently, economic models are useful because they help simplify complex situations so we can more easily understand how things fit together. Models also are of great use for policy purposes.

 

 

Positive and Normative Analysis

Health economists perform two types of analysis. Positive analysis uses economic theory and empirical analysis to make statements or predictions concerning economic behavior. It seeks to answer the question “What is?” or “What happened?” For example, we might investigate the exact relation between income and health care spending. Because positive analysis provides or predictions, it tends to be free of personal values.

Normative analysis, on the other hand, deals with the appropriateness or desirability of an economic outcome or policy. It seeks to answer the question “What ought to be?” or “Which is better?” For example, an analyst might conclude that households with incomes less than $30,000 per year should be subsidized by the government because they are unable to maintain a proper level of health care spending. Naturally, this implies that the analyst is making a value judgment. Because opinions vary widely concerning the desirability of any given economic outcome and the role government should play in achieving outcomes, it is easy to see why normative statements generally spark more controversy than positive ones. For instance, when 518 health economists were asked whether the Canadian health care system is superior to the U.S. system, there was much disagreement. Fifty-two percent of the economists agreed and 38 percent disagreed with the statement. The remaining 10 percent had no opinion or lacked the information needed to respond to the question (Feldman and Morrisey, 1990).

The following sets of positive and normative economic statements should give you a better understanding of the difference between the two. Notice that the positive statements deal with what is or what will be, whereas the normative statements concern what is better or what ought to be.

Positive: According to Becker and Murphy (1988), a 10 percent increase in the price of cigarettes leads to a 6 percent reduction in the number of cigarettes consumed.

Normative: The government should increase the tax on cigarettes to prevent people from smoking.

Positive: A study by Hellinger (1991) estimates that the average yearly cost of treating someone with AIDS is $38,300, while the lifetime costs equal $102,000.

Normative: It is in our country’s best interests that the federal government take a more active role in the prevention of AIDS.

Positive: National health care expenditures per capita are higher in the United States than Canada.

Normative: To control health care expenditures, the United States should adopt a national health insurance program similar to Canada’s.

 

 

Empirical Testing

Empirical testing of economic theories is important for two reasons. First, economic hypotheses require empirical validation, especially when a number of competing theories exist for the same real-world occurrence. For example, some people believe medical illnesses occur randomly whereas others believe medical illness is largely a function of lifestyle. The “random” and “lifestyle” explanations represent two competing theories for medical illnesses. Empirical studies can potentially ascertain which theory does a better job of explaining illnesses.

Second, even well-accepted theories are unable to establish the magnitude of the relation between any two variables. For example, suppose we accept the theory that lifestyle is a very important determinant of health status. A question remains about the magnitude or strength of the impact lifestyle has on health status. Does a young adult who adopts a sedentary lifestyle face a 10, 20, or 50 percent chance of dying prematurely compared to an otherwise comparable individual? Empirical studies can help provide the answer to that question.

There are many different ways for researchers to conduct an empirical analysis. The method we emphasize in this book, which most economists also use, is regression analysis. Regression analysis is a statistical method used to isolate the cause-and-effect relation among variables. Our goal is to provide the reader with an elementary but sufficient understanding of regression analysis so the regression results discussed in this book can be properly interpreted. Regression analysis is explained through an example.

The example used concerns the relation between health care expenditures, E, and consumer income, Y. Suppose we hypothesize that health care expenditures rise with household income and want to test our theory. Health care expenditures represent the dependent variable, and income is the independent variable. Furthermore, suppose we expect a linear (or straight-line) relationship between income and health care expenditures, or

 

(A1-4)                                    E = a + bY,

 

where a is the constant or intercept term and b is the slope parameter. If you recall, the slope parameter in this case identifies the change in health care expenditures that results from a one-unit change in income.

Because we are interested in the actual or real-world magnitudes of the parameters a and b, we will now collect a random sample of observations relating information on both medical expenditures and income. The data might be series observations on income and expenditures for a particular household over time or cross-sectional observations on income and expenditures across different households at a particular point in time. In this case, the household represents the unit of analysis but the unit of analysis could be an individual or a town, county, state, region, or country. Suppose we collect cross-sectional data on income and medical expenditures from a random survey of 30 households.

Exhibit A1-1 shows a scatter diagram illustrating our random sample of observations (only 5 of the 30 observations are illustrated for easier manageability). Notice that the scatter diagram of observations does not automatically show a linear relation between income and health care expenditures because of omitted factors that also influence spending on health care, some randomness to economic behavior, and measurement error. Our objective is to find the line that passes through those observations and provides the best explanation of the relation between Y and E. One can imagine numerous lines passing through the set of observations. What we want is the line that provides the best fit to the data.

A criterion is necessary to determine which line constitutes the best fit. One popular criterion is ordinary least squares, or OLS. OLS finds the best line by minimizing the sum of the squared deviations, e,, from the actual observations and a fitted line passing through the set of observations, or

 

(A1-5)                  Minimize 2 e² = 2 (E_a – Ef)² = 2 (E_a – a – b Y)²,

 

where E_a is the actual observation on medical expenditures and E_f is fitted (or predicted) expenditures from the estimated regression line, a + bY. In Exhibit A1-2, we show an example of a fitted line and the resulting deviations between actual and fitted expenditures. Based upon the sample of observations, a computer program (such as SAS, SPSS, or TSP) searches for the best line using the OLS procedure. In the process of finding the best line, the intercept and slope are determined, and thus we estimate the best magnitudes for a and b that minimize the sum of the squared deviations from the actual observations.

Let’s suppose the following results are obtained from the regression analysis:

(A1-6)                           E = 2,000 + 0.2Y.

 

A scatter diagram showing the actual relationship between household income and health care spending for 5 observations.

 

The results would tell us that the best fitted line to the data has an intercept of $2,000 and a slope of 0.2. Although the fitted or estimated regression line provides the “best” fit compared to all other lines, we do not know yet whether it represents a “good” fit to the actual data. Fortunately, the computer estimation procedure also provides us with some goodness-of-fit information that we can use to determine if the best fit is also a reasonably good one.

 

The fitted line resulting from OLS and the associated deviations between the fitted and actual values.

 

The two most common and elementary goodness-of-fit measures are the coefficient of determination, R², and the t-statistic, t. The coefficient of determination identifies the fraction of the variation in the dependent variable that is explained by the independent variable. Thus, the R² ranges between 0 and 1. Researchers tend to place more faith in a regression line that explains a greater proportion of the variation in the dependent variable.

The values for the parameters a and b are average estimates rather than true values because they are based on a sample instead of all possible observations; thus, they are associated with some error. Accordingly, there will be some deviations around the average estimate for a and also around the average estimate for b. In fact, if the deviations are very large, we cannot place much faith in the estimated value for the parameters. Indeed, the true value for b may be zero. If so, no relationship exists between income and health care expenditures.

The computed t-statistic helps us identify how much deviation occurs around the estimated average value for the parameters of the model. A t-statistic of 2 or more means that the value of the estimated parameter was at least twice as large as its average deviation. A rule of thumb is that when the t-statistic is 2 or more, we can place about 95 percent confidence in the estimated average value for the parameter, meaning that only a 5 percent likelihood exists that the relationship could have occurred by chance. Another rule of thumb is that when the t-statistic is 3 or more, we can place 99 percent confidence in our estimated value for the parameter. In this case, only a 1 percent likelihood exists that the relation occurred by chance.

Regression results are generally reported similar to the following:

 

(A1-7)                           E =   2,000 + 0.2Y       R² = 0.47

(2.52) (3.40)       N = 30

 

The t-statistics are reported in parentheses below the parameter estimates. Because the t-statistic associated with income is greater than 3, we can place a high degree of confidence in the parameter estimate of 0.2. Also, according to the regression results, income explains about 47 percent of the variation in health care expenditures. The number of observations, N, is 30.

Before we move on we need to interpret the parameter estimates for Equation A1-4. The intercept term of 2,000 tells us the level of health care expenditures when income is zero. The parameter estimate of 0.2 on the income variable is much more telling and suggests that expenditures on health care will increase by 20 cents if income increases by one-dollar. If the estimated parameter was instead -0.2, it would mean that a one-dollar increase in income causes health care expenditures to decrease by 20 cents. Thus, both the sign and value of the parameter estimate convey important information to the researcher.

The regression analysis we have been discussing thus far is an example of a simple regression because there is only one independent variable. Multiple regression refers to an analysis in which more than one independent variable is specified. For example, theory might tell us that price or tastes and preferences should also be included in an expenditure equation. The OLS procedure behind multiple regression is the same as that for simple regression and finds the best line that minimizes the squared deviations between the actual and fitted values. The computed R² identifies the variation in the dependent variable, say, health care expenditures, explained by the set of independent variables, which in our example would be price, income, and tastes and preferences. Each independent variable would be associated with an estimated parameter and t-statistic. For example:

 

(A1-8)                  E = 1,000 – 0.2P + 0.13Y + 0.8A    R² = 0.75

(2.32) (0.42) (3.23) (4.00)     N = 30

 

where P represents the price of medical services and A represents the average age in the household as a proxy for tastes and preferences. According to the regression results, the independent variables collectively explain 75 percent of the variation in health care expenditures. Also, the regression results suggest that income and age both have a statistically significant direct impact on health care expenditures. Price, on the other hand, has no impact on health care expenditures according to the regression findings.

 

 

Association versus Causation

As mentioned previously, the intent behind multiple regression analysis is to establish a cause and effect relationship among variables. Sometimes, however, multiple regression analysis simply captures an association or correlation among variables rather than a true causal relationship. That happens most often for observational studies that involve crosssectional or time series data but contaio correction for the circumstances behind the observed relationship. The association but lack of causation typically occurs because the underlying observations have not resulted from a randomized process with both a control and a treatment group. Figure A1-3 helps to show why an observational study may be hindered by its inability to distinguish between a causal relationship and an association.

Office visits are associated with physical health status because an unobservable factor, such as mental health status, affects both.

 

The figure illustrates a simple relationship between physical health status (say a selfreported index ranging from poor to excellent physical health) and the number of physician visits (as a measure of medical care). All other measurable factors affecting physical health status, like age, gender, and income, are collapsed and captured in the variable X. Suppose we are investigating if more office visits help to improve, or cause, better health. However, even if the multiple regression analysis yields a statistically significant relation between the number of physician visits and more favorable health we cannot be certain if the evidence supports a causal relationship. The uncertainty holds for two reasons.

First, a third unobservable and therefore immeasurable factor, Z, that cannot be included in X, may simultaneously affect both the number of physician visits and physical health status and thereby produce the observed association. For example, suppose we cannot properly and completely measure mental health status (e.g., the severity of depression) and mental health status influences both the self-reported physical health index and the likelihood of visiting a physician. Perhaps, severely depressed individuals simultaneously downgrade their physical health status and become more reclusive so they fail to visit their physician. If so, any observed correlation between physician visits and physical health status, in the presence of this important omitted unobservable variable, may not reflect causation.

Second, reverse causality may pose a problem when attempting to draw inferences about the direction of causal relationships from regression results. That is, physical health status, the dependent variable in Figure A1-3, may influence the number of physician visits, the independent variable. For example, state governments may pursue policies to encourage more doctors per person in areas with the highest infant mortality rates. Or, pregnant mothers may be more likely to seek out physicians when they suspect the health of their infants may be at greater risk. Hence, the regression results from an observational study would actually reflect a reverse effect – health status causes visits.

As a result, investigators often use various methods to identify or isolate causal relationships. Basically, some type of identification strategy is necessary to distinguish a causal relationship from an association. One strategy randomly assigns people or households to different situations or categories and conducts a controlled behavioral experiment. Following our same example, on a random basis, various individuals might be required to visit the doctor a certaiumber of times per year. Some individuals may not be allowed any physician visits at all, and others may be forced to visit their doctor ranging from one to ten times per year, regardless of their income, observable mental health status, or other personal characteristics. The random assignment of households corrects for any self-selection bias that results when individuals with different (unobservable) mental health states are allowed to choose the number of doctor visits.

The analyst then studies the relation between the number of office visits and physical health status, while controlling for other observable measures that may also affect health status using a technique such as multiple regression analysis. The hypothesis is that physical health status improves with more office visits – ceteris paribus. As you might expect, randomized social experiments of this kind offer valuable insights but are very expensive to conduct. In addition, the health of some individuals might be seriously compromised if they are not permitted to visit the doctor a reasonable number of times per year. Hence large social experiments are rarely conducted. In Topic 5, we will discuss the RAND Health Insurance Study of the 1970s which randomly assigned households to different health plans and investigated various hypotheses relating to health and health care.

A natural experiment, an alternative identification strategy, arises when some type of external global policy, unrelated to other determinants of physical health status, produces an uncontrollable shock in the medical care received by a treatment group. Changes in the health outcomes of this treatment group are then compared to health outcomes of the control group that did not experience that same external shock but otherwise faced fairly similar circumstances. The uncontrollable nature of the policy shock prevents self-selection.

For example, suppose the government sharply cuts funding for various public health insurance plans such that some low-income people are randomly terminated from the programs. Those individuals terminated from the programs represent the treatment group and those continuing in the programs represent the control group. After a given period, we then gather data on physical health status and other determinants of health status including age, gender, and income.

In the multiple regression analysis, physical health status serves as the dependent variable. The independent variables include a 0 or 1 dummy variable identifying if the individual was subjected to the policy shock or not, and other measurable determinants of physcial health status. Assuming 1 represents an individual in the treatment group, we would expect a negative coefficient on the dummy variable because termination from the programs causes poorer health, all other factors held constant.

Several natural experiments have studied the effect of medical care program terminations (such as veteran or maternal health benefits) on the health outcomes of a treatment group compared to an otherwise similar control group for which the termination did not occur (Levy and Meltzer, 2001). While this method offers a valuable way of identifying the existence of causal relationships, various drawbacks exist. First of all, not many policy shocks occur in practice for testing various hypotheses. Even when they do, the so-called treatment and control groups may not be randomly selected. For example, in some of the studies just cited, only those individuals with less severe illnesses were terminated from the medical care programs.

The third identification strategy is called the instrumental variables approach. To conduct the instrumental variables approach, in the context of our example, a variable (i.e., an instrument) or a set of variables that affect the number of office visits but not physical health status must be found. For instance, the distance of each household from the physician’s office might be used as an instrument because it could be argued that distance helps to determine the number of office visits (i.e., convenience), but not physical health status.

If so, a multiple regression technique called two stage least squares can be employed to examine the extent to which distance affects the number of physician visits in the first stage of the estimation procedure and then the effect of physician visits on physical health status in the second stage. This technique essentially purges some of the association between physician visits and physical health status resulting from the third variable problem or reverse causality. That is, we can identify any change in physical health that results from a change in the number of office visits because of less or greater convenience.

The instrumental variables approach is one of the more popular methods for identifying causal relationships. However, in practice, it is often difficult to find a suitable set of instruments. This is particularly true for health economic analyses where many variables are highly correlated with one another such as the consumption of medical care, health insurance status, income, and health status – it is very hard to find a factor of set of factors that affect one but not the others.

The final method to identify a causal relationship is referred to as the fixed effects model. A panel data set, which combines both cross-sectional and time series data, is necessary to use a fixed effects model. The same 100,000 people over 10 years or 50 states over 20 years represent examples of panel data sets. Because of the time dimension, we can track how the same cross-section of observations reacts to changes in various factors over time. More importantly, a 0/1 dummy variable for each cross-section observation in the sample can be specified in the multiple regression equation to control for unobservable heterogeneity (i.e., unobservable differences among the cross-section observations).

Recall from our running example that we are unable to control for the severity of mental depression and that omitted variable creates a third variable problem. Assuming each individual’s state of mental depression is fairly constant over time, the set of cross-section dummy variables or fixed effects essentially helps to control for mental health status differences as well as any other unobservable differences among the individuals in the sample. This reduces the likelihood of a third variable problem and allows the researcher to better identify a causal relationship.

For that reason, most of the statistical research today in health economics involves a fixed effects model. There are a couple of shortcomings associated with the fixed effects approach, however. First, data requirements are much greater. Data for the same crosssection of observations must be obtained and inputted for a number of years. But with greater amounts of data available on-line and in predetermined formats, that shortcoming is becoming less troublesome. Second, the fixed effects model assumes that the unobservable heterogeneity, e.g., severity of mental depression, is relatively constant over time. If the unobservable variable changes over time, then the third variable problem may not be eliminated and the empirical results may reflect an association instead of a causal relationship. When a social or natural experiment cannot be performed, a preferred identification strategy combines an instrumental variables approach along with a fixed effects model.

 

 

What Is Health?

The Mosby Medical Encyclopedia (1992, p. 360) defines health as “a state of physical, mental, and social well-being and the absence of disease or other abnormal condition.” Economists take a radically different approach. They view health as a durable good, or type of capital, that provides services. The flow of services produced from the stock of health “capital” is consumed continuously over an individual’s lifetime (see Grossman, 1972a, 1972b). Each person is assumed to be endowed with a given stock of health at the beginning of a period, such as a year. Over the period, the stock of health depreciates with age and may be augmented by investments in medical services. Death occurs when an individual’s stock of health falls below a critical minimum level.

Naturally, the initial stock of health, along with the rate of depreciation, varies from individual to individual and depends on many factors, some of which are uncontrollable. For example, a person has no control over the initial stock of health allocated at birth, and a child with a congenital heart problem begins life with a below-average stock of health. However, we learn later that medical services may compensate for many deficiencies, at least to some degree. The rate at which health depreciates also depends on many factors, such as the individual’s age, physical makeup, lifestyle, environmental factors, and the amount of medical care consumed. For example, the rate at which health depreciates in a person diagnosed with high blood pressure is likely to depend on the amount of medical care consumed (is this person under a doctor’s care?), environmental factors (does he or she have a stressful occupation?), and lifestyle (does the person smoke or have a weight problem?). All these factors interact to determine the person’s stock of health at any point in time, along with the pace at which it depreciates.

Regardless of how you define it, health is a nebulous concept that defies precise measurement. In terms of measurement, health depends as much on the quantity of life (that is, number of life-years remaining) as it does on the quality of life. Quality of life has become an increasingly important issue in recent years due to the life-sustaining capabilities of today’s medical technology. The issue gained national prominence in 1976 when, in a landmark court decision, the parents of Karen Ann Quinlan were given the right to remove their daughter, who was in a persistent vegetative state, from a ventilator. Because the quality of life is a relative concept that is open to wide interpretation, researchers have wrestled with developing an instrument that accurately measures health. In Topic 3, we will discuss some of these measures.

 

 

Why Good Health? Utility Analysis

As mentioned earlier, health, like any other durable good, generates a flow of services. These services yield satisfaction, or what economists call utility. Your television set is another example of a durable good that generates a flow of services. It is the many hours of programming, or viewing services, your television provides that yield utility, not the set itself.

As a good, health is desired for consumption and investment purposes. From a consumption perspective, an individual desires to remain healthy because she or he receives utility from an overall improvement in quality of life. In simple terms, a healthy person feels great and thus is in a better position to enjoy life. The investment element concerns the relation between health and time. If you are in a positive state of health, you allocate less time to sickness and therefore have more healthy days available in the future to work and enhance your income or to pursue other activities, such as leisure. Economists look at education from the same perspective. Much as a person invests in education to enhance the potential to command a higher wage, a person invests in health to increase the likelihood of having more healthy days to work and generate income.

The investment element of health can be used to explain some of the lifestyle choices people make. A person who puts a high value on future events is more inclined to pursue a healthy lifestyle to increase the likelihood of enjoying more healthy days than a person who puts a low value on future events. A preference for the future explains why a middle-aged adult with high cholesterol orders a salad with dressing on the side instead of a steak served with a baked potato smothered in sour cream. In this situation, the utility generated by increasing the likelihood of having more healthy days in the future outweighs the utility received from consuming the steak dinner. In contrast, a person who puts a much lower value on future events and prefers immediate gratification may elect to order the steak dinner and ignore the potential ill effects of high cholesterol and fatty foods.

Naturally, each individual chooses to consume that combination of goods and services, including the services produced from the stock of health, which provides the most utility. The isolated relation between an individual’s stock of health and utility is captured in Figure 2-1, where the quantity of health, H, is measured on the horizontal axis and the level of utility, U, is represented on the vertical axis. To simplify matters, we ignore the intermediate step between the health stock, the services it provides, and the utility received from these services and assume that the stock of health directly yields utility. The positive slope of the curve indicates that an increase in a person’s stock of health directly enhances total utility. The shape of the curve is particularly important because it illustrates the fundamental economic principle of the law of diminishing marginal utility. This law states that each successive incremental improvement in health generates smaller and smaller additions to total utility; in other words, utility increases at a decreasing rate with respect to health.

 

 

The total utility curve is upward sloping and depicts the relation between an individual’s stock of health and utility. The positive slope indicates that total utility increases as an individual’s stock of health improves; the bowed shape of the curve captures the impact of the law of diminishing marginal utility. This law is a fundamental principle of economics stating that each additional improvement in health generates an ever smaller increase in utility. Notice that the increase in health from H₀ to H1 causes utility to increase from U_Q to while an equal increase in health for H₂ to H3 results in a smaller increase in utility from U₂ to U₃.

 

For example, in Figure 2-1 an increase in health from H₀ to H₁ causes utility to increase from U₀ to U₁, while an equal increase in health from H₂ to H₃ generates a much smaller increase in utility, from U₂ to U₃. In the second case, the increase in utility is less when the stock of health is greater because of the law of diminishing marginal utility. The implication is that a person values a marginal improvement in health more when sick (that is, when having a lower level of health) than when healthy. This does not mean every individual derives the same level of utility from a given stock of health. It is possible for two or more people to receive a different amount of utility from the same stock of health. The law of diminishing marginal utility requires only that the addition to total utility decreases with successive increases in health for a given individual.

Another way to illustrate the law of diminishing marginal utility is to focus on the marginal utility associated with each unit of health. Marginal utility equals the addition to total utility generated by each successive unit of health. In mathematical terms,

 

(2-1)                                       MU_H = ΔU/ΔH,

 

where MU_H equals the marginal utility of the last unit of health consumed and D represents the change in utility or health. In Figure 2-1, Equation 2-1 represents the slope of a tangent line at each point on the total utility curve. The bowed shape of the total utility curve implies that the slope of the tangent line falls as we move along the curve, or that MU_H falls as health increases.

 

The MU curve illustrates the relation between marginal utility and the stock of health, and it is downward sloping because of the law of diminishing marginal utility. The shape of curve reflects the notion that each additional improvement in health results in a smaller increase in utility than the previous one.

 

Figure 2-2 captures the relation between marginal utility and the stock of health. The downward slope of the curve indicates the law of diminishing marginal utility holds because each new unit of health generates less additional utility than the previous one.

 

 

What Is Medical Care?

Medical care is composed of myriad goods and services that maintain, improve, or restore a person’s health. For example, a young man might have shoulder surgery to repair a torn rotator cuff so that he can return to work, an elderly woman may have hip replacement surgery so she can walk without pain, or a parent may bring a child to the hygienist for an annual teeth cleaning to prevent future dental problems. Prescription drugs, wheelchairs, and dentures are examples of medical goods, while surgeries, annual physical exams, and visits to physical therapists are examples of medical services.

Because of the heterogeneous nature of medical care, units of medical care are difficult to measure precisely. Units of medical care are also hard to quantify because most represent services rather than tangible products. As a service, medical care exhibits the four Is that distinguish it from a good: intangibility, inseparability, inventory, and inconsistency (Berkowitz et al., 1989).

The first characteristic, intangibility, means that a medical service is incapable of being assessed by the five senses. Unlike a new car, a steak dinner, or a new CD, the consumer cannot see, smell, taste, feel, or hear a medical service.

Inseparability means that the production and consumption of a medical service take place simultaneously. For example, when you visit your dentist for a checkup, you are consuming dental services at the exact time the dentist is producing them. In addition, a patient often acts as both producer and consumer. Without the patient’s active participation, the medical product is likely to be poorly produced. Educational services, like medical services, require the consumer’s active participation; that is, education is likely to be poorly provided when the student plays a passive role in the process.

Inventory is directly related to inseparability. Because the production and consumption of a medical service occur simultaneously, health care providers are unable to stockpile or maintain an inventory of medical services. For example, a dentist cannot maintain an inventory of dental checkups to meet demand during peak periods.

Finally, inconsistency means that the composition and quality of medical services consumed vary widely across medical events. Although everyone visits a physician at some time or another, not every visit to a physician is for the same reason. One person may go for a routine physical, while another may go because he needs heart bypass surgery. The composition of medical care provided or the intensity at which it is consumed can differ greatly among individuals and at different points in time.

The quality of medical care is also difficult to measure. Quality differences are reflected in the structure, process, and/or outcome of a medical care provider (Donabedian, 1980, 1988). Structural quality is reflected in the physical and human resources of the medical care provider, such as the facilities (level of amenities), medical equipment (type and age), personnel (training and experience), and administration (organization structure). Process quality reflects the specific actions health care providers take on behalf of patients in delivering and following through with care. Process quality might include access (waiting time), data collection (background history and testing), communication with the patient, and diagnosis and treatment (type and appropriateness). Outcome quality refers to the impact of care on the patient’s health and welfare as measured by patient satisfaction, work time lost to disability, or postcare mortality rate. Because it is extremely difficult to keep all three aspects of quality constant for every medical event, the quality of medical services, unlike that of physical goods, is likely to be inconsistent.

As you can see, medical care services are extremely difficult to quantify. In most instances, researchers measure medical care in terms of either availability or use. If medical care is measured on an availability basis, such measures include the number of physicians or hospital beds available per 1,000 people. If medical care is measured in terms of use, the analyst employs data indicating how often a medical service is actually delivered. For example, the quantity of office visits or surgeries per capita is often used to represent the amount of physician services rendered, whereas the number of inpatient days is frequently used to measure the amount of hospital or nursing home services consumed.

 

 

The Production of Good Health

Health economists take the view that the creation and maintenance of health involves a production process. Much as a firm uses various inputs, such as capital and labor, to manufacture a product, an individual uses medical inputs and other factors, such as a healthy lifestyle, to produce health. The relation between medical inputs and output can be captured in what economists call a production function. A health production function indicates the maximum amount of health that an individual can generate from a specific set of inputs in a given period of time. In mathematical terms it shows how the level of output (in this case, health) depends on the quantities of various inputs, such as medical care. A generalized short-run health production function for an individual takes the fol­lowing form:

 

(2-2)             Health = H (medical care, technology, profile,

                                      lifestyle, socieconomic status, environment )

 

where health reflects the level of health at a point in time; medical care equals the quantity of medical care consumed; technology refers to the state of medical technology at a given point in time; profile captures the individual’s mental and physical profile as of a point in time; lifestyle represents a set of lifestyle variables, such as diet and exercise; socio-econonomic status reflects the joint effect of social and economic factors, such as education, income and poverty; and environment stands for a variety of environmental factors, including air and water quality.

 

The total product curve is upward sloping and indicates that as an individual consumes more medical care, overall health improves. The positive intercept term represents the individual’s level of health wheo medical care is consumed and is a function of other factors such as lifestyle and the environment. The law of diminishing marginal productivity accounts for the bowed shape of the curve. This law is a fundamental principle of production theory and it implies that health increases at a decreasing rate when additional units of health care are consumed, holding all other inputs in the health production process constant.

 

To focus on the relation between health and medical care, we assume initially that all other factors in the health production function remain constant. Figure 2-3 depicts this relation, where q is a hypothetical measure of medical care, holding technology constant, and H represents the level of health. The intercept term represents the individual’s level of health when zero medical care is consumed. As drawn, the total product curve implies that an individual’s level of health is positively related to the amount of medical care consumed. However, we should not rule out the possibility that poor health status or an illness might be created by additional medical services. An illness created by a medical care encounter is referred to as an iatrogenic disorder, “a condition caused by medical personnel or procedures or through exposure to the environment of a health-care facility” (Mosby Medical Encyclopedia, p. 401). For example, a physician may accidentally harm a patient by prescribing the wrong medicine for a given medical condition. The shape of the curve is very similar to that in Figure 2-1 and reflects the law of diminishing marginal productivity. This law implies that health increases at a decreasing rate with respect to additional amounts of medical care, holding other inputs constant. For example, suppose an individual makes an initial visit and several follow-up visits to a physician’s office for a specific illness or treatment over a given period of time. It is very likely that the first few visits have a more beneficial impact on the individual’s stock of health than the later visits. Thus, each successive visit generates a smaller improvement in health than the previous one.

 

The MP curve establishes the relation between the marginal product of medical care and the amount of medical care consumed. The curve is downward sloping because the marginal product of the last unit of medical care consumed decreases as the individual consumes more medical care, reflecting the law of diminishing marginal productivity.

 

The relation between health and medical care can also be viewed from a marginal perspective, where the marginal product of medical care represents the incremental improvement in health brought about by each successive unit of medical care consumed, or

 

(2-3)                                       MP_q = ΔH/Δq,

 

where MP_q equals the marginal product of the last unit of medical care services consumed. The law of diminishing marginal productivity holds that the marginal product of medical care diminishes as the individual acquires more medical care. A graph of this relationship appears as a negatively sloped curve in Figure 2-4. As in utility analysis, the marginal product of medical care equals the slope of a tangent line drawn to every point on Figure 2-3.

The other variables in the health production function can also be incorporated into the analysis. In general terms, a change in any one of the other variables in the production function alters the position of the total product curve. The total product curve may shift in some instances and/or rotate in others. In the latter case, the curve rotates because the marginal productivity of medical care has changed in response to the change in the other factors.

New medical technologies have profoundly affected all aspects of the production of medical care. In the broadest of terms, examples of new technologies include the development of sophisticated medical devices, the introduction of new drugs, the application of innovative medical and surgical procedures, and most recently, the use of computer-supported information systems, just to name a few. According to Cutler and Huckman (2003) and Cutler and McClellan (2001), technological change can result in treatment expansion, treatment substitution, or some elements of both. Treatment expansion occurs when more patients are treated by a new medical intervention, perhaps because of a higher success rate or lower risks to health. Treatment substitution occurs when the new technology substitutes for or replaces an older one.

In the context of our health production model, the development and application of a new medical technology causes the total product curve to pivot and rotate upward because the marginal productivity of each unit of medical care consumed increases, as illustrated in Figure 2-5. Notice that the total product curve rotates upward from TP₀ to TP₁ and each unit of medical care consumed now generates a greater amount of health. The movement from point A to point B in Figure 2-5 illustrates the case in which the improvement in medical technology brings about an increase in the amount of medical care consumed from q₀ to q₁ along with an improvement in health from H₀ to H₁. This movement represents the treatment expansion resulting from the new medical technology. Movement from point A to point C illustrates the situation in which the new technology has no impact on health but results in less consumption of medical care from q₀ to q₂. In this case, the new technology is cost saving, everything else held constant. It should be noted that the increase in the marginal product of medical care brought about by the medical technology also causes the marginal product curve to shift to the right.

 

The total product curve shifts upward with the development and application of new medical technology because of an increase in the marginal product of medical care. A movement from point A to point B illustrates the case in which a new technology results in a simultaneous increase in the amount of medical care consumed and improvement in health. A movement from point A to point C depicts the case in which the new medical technology has no impact on health but results in less consumption of medical care.

 

 

The profile variable in Equation 2-2 depends on a host of variables and controls for such items as the person’s genetic makeup, mental state, age, gender, and race/ethnicity as of a given point in time (such as the beginning of the year). Any change in the profile variable affects both the intercept term and the slope of the health production function. For example, an individual’s genetic makeup may make him or her a candidate for prostate or breast cancer. If this individual gets cancer for that reason, then his or her total product curve shifts downward. That is because overall health has decreased regardless of the amount of medical care consumed. The total product curve is also likely to rotate downward at the same time because the marginal product of medical care should decrease as the profile worsens. The total product curve rotates downward because an otherwise healthy person is likely to respond more favorably to medical treatments for a given medical complication than one who is less healthy. Both of these changes are represented in Figure 2-6, where the total product curve shifts and rotates downward at the same time from TP₀ to TP₁. The marginal product curve for medical services also shifts to the left, because each incremental unit of medical care now brings about a smaller improvement in health.

The effect of age on the production of health is relatively straightforward. Age affects health through the profile variable. As an individual ages and deteriorates physically, both health and the marginal product of medical care are likely to fall. In addition, the rate at which health depreciates over the period is also likely to increase with age. This causes the total product curve to shift downward and flatten out. The decrease in the marginal product of medical care also causes the marginal product curve to shift to the left. The impact of gender on the total and marginal product curves is left to the reader and is the focus of a review question at the end of this topic.

The graph illustrates what happens to the total product curve when an individual gets an illness such as cancer for a reason other than improper medical care. The curve shifts downward because at each level of medical care consumed the individual is less healthy than previously was the case. The curve also rotates downward and becomes flatter, reflecting the likelihood that the now ill individual is going to respond less favorably to a given amount of medical care consumed, such as an office visit, than previously was the case when she was healthy.

However, we cannot rule out the reverse effect that health influences education, par­ticularly during childhood. Take the case of a child with chronic asthma where an asthma attack can be brought on by any number of events such as exposure to allergies or viral infections, and physical exertion. As a result, a child with chronic asthma is more likely to miss school, learn less while attending school, and in the end acquire less education. Over time, what the researcher may observe is a less healthy adult with only a modest level of education.

Lifestyle variables consider the impact of personal health habits on the production of health. Personal habits include such things as whether the person smokes, drinks excessively, leads a sedentary lifestyle, is overweight, or has an improper diet. For example, consider a newly health-conscious individual who decides that a change in lifestyle is in order. After a regimen of diet and exercise, this person loses some weight and improves his or her physical conditioning. As a result of this change in lifestyle, the individual’s level of health and the marginal product of medical care should increase. This causes the total product curve to shift and rotate upward.

As is the case with improvements in personal habits, improved socioeconomic conditions cause the intercept term and the marginal product of medical care to increase. For example, since education is likely to make the individual a more efficient producer of health independently of the amount of medical care consumed, the total product curve shifts upward. An individual with more education is likely to better understand the positive impact of a healthy diet on health. The total product curve also steepens, or the marginal product of medical care increases, because education allows the person to utilize each unit of medical care consumed more effectively. For example, an educated individual may be more inclined to understand and follow a physician’s advice concerning diet and exercise after undergoing a heart bypass operation. In addition, she or he may be able to recognize a medical problem early and seek medical care quickly when the effectiveness of medical treatment is generally at its maximum.

 

The graph illustrates what happens to the total product curve when an individual gets an illness such as cancer for a reason other than improper medical care. The curve shifts downward because at each level of medical care consumed the individual is less healthy than previously was the case. The curve also rotates downward and becomes flatter, reflecting the likelihood that the now ill individual is going to respond less favorably to a given amount of medical care consumed, such as an office visit, than previously was the case when she was healthy.

 

Some analysts have hypothesized that the relation between education and health is far more complex. For example, Fuchs (1979) argues that the acquisition of education and health depends on the value people place on future events, or the rate at which they discount future events. Individuals who place a high value on future benefits and are willing to postpone gratification are inclined to acquire more education and pursue a healthier lifestyle when they are young. This is because they want to reap the rewards of a higher income and a longer life that more education and a healthier lifestyle can bring. On the other hand, individuals who place a low value on future events and desire immediate gratification are not likely to acquire significant amounts of education or to follow a healthy lifestyle because they have adopted a “live for today” attitude. Thus, according to Fuchs, higher levels of education may be associated with better health not because there is a direct link between the two variables but because both variables are directly correlated with a third factor, the degree to which future events are valued.

The impact of income on health is also complex and is referred to as the “income gradient” in the literature “to emphasize the gradual relationship between the two: health improves with income throughout the income distribution” (Deaton, 2002, p. 14). Income is likely to indirectly impact health through a number of pathways. An increase in income provides the individual the means to consume more medical care. In addition, a more affluent individual is likely to be more educated, pursue a healthier lifestyle, and live in a safer environment, all of which contribute to improved health. For example, a more affluent individual may live in a suburban community where the crime rate is low, access to drugs and alcohol is limited, and quality medical care is available just around the corner. Income may also have a direct impact on health, although the net effect is far from clear. On the one hand, a wealthier individual may be employed in a safer work environment where the risk of a work-related accident or illness is slim. On the other hand, a wealthier individual may be employed in a more stressful occupation, which can adversely impact health.

In recent years an extensive body of literature has developed that examines whether the distribution of income impacts health, and the income-health hypothesis has taken on a variety of forms. According to the literature (Lynch et al., 2004; Wagstaff and van Doorslaer, 2000), the various hypotheses that have been offered over time can be classified into four broad categories: the absolute income hypothesis, the relative income or deprivation hypothesis, the relative position hypothesis, and the income inequality hypotheses.

The absolute income hypothesis simply states that an individual’s absolute income is positively related to health for the reasons discussed previously. The relative income or deprivation hypothesis posits that an individual’s income relative to some social group average impacts overall health. Put in more definable terms, it is a person’s income relative to some critical level such as the poverty line in the United States that matters. The presumption is that anyone with an income below the poverty line lacks the ability to acquire the basic necessities, such as health care.

The relative position hypothesis emphasizes that one’s social position in the income distribution also impacts health. For example, those at the bottom of the income scale in the United States may become frustrated and feel left behind by the “American dream” despite the fact that they have enough income to live in reasonable housing and receive adequate health care. Out of a sense of discouragement, these people may tend to give up and pursue a lifestyle detrimental to their health that could involve increased alcohol consumption, smoking, and obesity.

Finally, the income inequality hypothesis states that the distribution of income itself directly impacts health. For example, greater income inequality may create an incentive for government to limit spending on social programs that have a direct bearing on health in an attempt to lower taxes. Greater income inequality may also lead to an erosion of social capital, defined as “those features of social organizations – such as the extent of interpersonal trust between citizens, norms of reciprocity, and vibrancy of civic organizations – that facilitate cooperation for mutual benefit” (Kawachi and Kennedy, 1999, p. 221). As a result, the poor may find their public health needs largely ignored by society at large.

An adjustment in a person’s physical environment is also likely to affect the total product curve. For example, an individual with an asthmatic condition might move from Los Angeles, where smog is intense, to a community on the far outskirts of the city. Or the person’s spouse may give up smoking to decrease the level of secondhand smoke in the home. As a result, the probability that this person will succumb to a respiratory ailment diminishes. Both of these changes cause the total product curve to shift and rotate upward.

In short, health production theory suggests that a variety of factors, such as the individual’s profile, medical care, state of medical technology, lifestyle, socioeconomic status, and environment, interact to determine health. The theory also suggests that health increases at a diminishing rate with respect to greater amounts of medical care consumed, provided all other inputs remain constant. If any other inputs in the production process change, the impact of medical care on health is also likely to change. The effect of any one nonmedical input on health is also likely to exhibit diminishing returns – all other inputs held constant. For example, running two miles a day may reduce someone’s weight by 15 pounds over a six-month period. It is doubtful, however, that an additional two miles per day of running could produce additional 15 pounds of weight loss during the next six-month period.

Before we conclude this section, you should be aware that recently Jacobson (2000), Bolin et al. (2002), and others have extended the Grossman model and developed a number of sophisticated mathematical models that focus on the family rather than the individual as the main producer of health. While these models are beyond the scope of this book, they represent a valuable addition to the literature. The common theme is that individual decisions to invest in health are made within the context of a family and that any decision on the part of one family member regarding investments in health impacts the health investment decisions of others in the family. For example, a learning-disabled child may provide an incentive to a mother to invest more in her own health to ensure that she will have the time to aid her child. These theoretical developments provide a number of challenges to researchers as they strive to understand the complex relationships between family members and individual health-related decisions.

 

Summary

Economic models and empirical testing of hypotheses are important for making sense of the real world, for advancing knowledge, and for public policy purposes. Economic models help to organize our thoughts about the relationship among key variables by helping to simplify an otherwise complex world. Positive analysis cannot be performed without economic models and normative analysis should be based on solid positive theory.

Empirical evidence should also be based on sound economic theory. That is, the variables specified in a multiple regression equation should be based on economic reasoning rather than ad hoc notions. Knowing the quantitative magnitude of the relationships among variables provides important insights into the relative effectiveness of various policies. As a result, choosing the best policy often requires hard empirical evidence.

We recognize that learning the material in this appendix does not make the reader an econometrician. Econometrics is way too complex for that to happen. The material does, however, introduce the reader to the general idea behind the empirical testing of health economic hypotheses. It also exposes the reader to some of the pitfalls involved and several techniques for dealing with these pitfalls. The basic idea is that all multiple regression models are not created equally; some are clearly better than others. We invite you to learn more about the theory and practice of econometrics. The website for the text at http://www.cengage.com/economics/santerre contains another more formal econometric appendix written by Bruce Carpenter of Mansfield University. It goes into great detail on the specifics behind multiple regression analysis, logarithmic functions, and how elasticities can be determined with the estimated coefficients among other topics. Studenmund (2006) offers a good introduction to econometric issues. Also, Dowd and Town (2002) offer a worthwhile discussion of causation versus association.

 

 

Review Questions and Problems

1.                Determine whether the following statements are based on positive or normative analysis. Be sure to substantiate your answers.

A.                 Prices of physician services should be controlled by the government because many citizens cannot afford to pay for a visit to a physician.

B.                  According to Tosteson et al. (1990), a 25 percent drop in the number of people who smoked in 1990 would reduce the incidence of coronary heart diseases by 0.7 percent by the year 2015.

C.                 Rising health care costs have forced numerous rural hospitals to close their doors in recent years.

D.                 According to government statistics, in 1989 7.2 deaths per 100,000 residents were alcohol induced. To decrease this number, the government should impose higher taxes on alcohol.

2.                Suppose a health expenditure function is specified in the following manner:

E = 500 + 0.2Y,

where E represents annual health care expenditures per capita and Y stands for income per capita.

A.                 Using the slope of the health expenditure function, predict the change in per capita health care expenditures that would result from a $1,000 increase in per capita income.

B.                  Compute the level of per capita health care spending when per capita income takes on the following dollar values: 0; 1,000; 2,000; 4,000; and 6,000.

C.                 Using the resulting values for per capita health care spending in part B, graph the associated health care expenditure function.

D.                 Assume that the fixed amount of health care spending decreases to $250. Graph the new and original health care functions on the same graph. What is the relation between the original and new health care expenditure functions?

E.                  Now assume that the fixed amount of health care spending remains at $500 but the slope parameter on income decreases to 0.1. Graph both the original and new health care expenditure functions. Explain the relation between the two lines.

3.                Victor Fuchs (1996) lists the following questions in an article in The Wall Street Journal. Identify whether the following questions involve positive or normative analysis. All the questions deal with a Republican plan to reform Medicare, the public health insurance program for the elderly.

A.                 How many Medicare beneficiaries will switch to managed care?

B.                  How much should the younger generation be taxed to pay for the elderly?

C.                 Should seniors who use less care benefit financially, or should they subsidize those who use more care?

D.                 How many Medicare beneficiaries will switch to medical savings accounts (see Topic 16)?

E.                  What effect will these changes have on utilization?

F.                  How much should society devote to medical interventions that would add one year of life expectancy for men and women who have already passed the biblical “three score and ten”?

G.                 Will senior citizens’ choices about types of coverage depend on their health status?

H.                 If the rate of spending growth is reduced to 6 percent from 10 percent a year, what will happen to the growth of medical services? To physician incomes?

4.                Indentify two purpose of empirical testing.

5.                Suppose you are explaining the technique behind OLS to a statistically-challenged but otherwise intelligent uncle of yours. Further suppose the statistical relationship concerns one between the number of physician visits and physical health status. Don’t worry about drawing causality but only explaining the OLS technique itself. Explain to him how OLS fits a line to a set of observations. You might want to use a scatter diagram and an equation for a line to make your point.

6.                Suppose you are presented with the following regression equation involving health care expenditures and its determinants, where all of the variables have been defined previously.

 

E = 500 – 25P + 0.20Y – 1.2A                  R² = 0.30

(1.21) (2.45) (0.43) (4.13)      N = 1,000

 

a.                   What percent of the variation in health care spending is explained by the various independent variables?

b.                  Which of the independent variable possess a statistical significant impact on health care spending? What do the results suggest about the relation between income and health care spending?

c.                   Supposing that both P and E are measured in dollars, interpret the coefficient estimate on P.

d.                  What does the coefficient estimate on A suggest about the relation between age and health care spending?

e.                   Can you think of any omitted variables that might cause our estimates to be suspect?

7.                Some years ago several researchers found a correlation between cigarette smoking and suicides. Do you think this correlation reflects an association or a causal relationship? Why? If it reflects an association, can you think of a plausible third variable?

8.                What are meant by the third variable problem and reverse causation?

9.                In your own words, explain the difference between a social experiment and a natural experiment.

10.           In your own words, explain how the instrumental variables and fixed effects approaches deals with the third variable problem.

 

 

References

Becker, Gary S., and Kevin M. Murphy. “A Theory of Rational Addiction.” Journal of Political Economy 96 (August 1988), pp. 675-700.

Dowd, Bryan, and Robert Town. “Does X Really Cause U?” AcademyHealth, Washington, D.C., September 2002.

Feldman, Roger, and Michael A. Morrisey. “Health Economics: A Report on the Field.” Journal of Health Politics, Policy and Law 15 (fall 1990), pp. 627-46.

Fuchs, Victor R. “The Tofu Triangle.” The Wall Street Journal, January 26, 1996, p. A16.

Hellinger, Fred J. “Forecasting the Medical Care Costs of the HIV Epidemic: 1991-1994.” Inquiry 28 (fall 1991), pp. 213-25.

Levy, Helen, and David Meltzer. “What Do We Really Know about Whether Insurance Affects Health”. Mimeo, University of Chicago, 2004.

Studenmund, A.H., Using Econometrics: A Practical Guide, Addison-Wesley, 2006.

Tosteson, Anna, et al. “Long-Term Impact of Smoking Cessation on the Incidence of Coronary Health Disease.” American Journal of Public Health 80 (December 1990), pp. 1481-86.